[PATCH] x86_64: When user could have changed RIP always force IRET

Intel EM64T CPUs handle uncanonical return addresses differently
from AMD CPUs.

The exception is reported in the SYSRET, not the next instruction.
This leads to the kernel exception handler running on the user stack
with the wrong GS because the kernel didn't expect exceptions
on this instruction.

This version of the patch has the teething problems that plagued an earlier
version fixed.

This is CVE-2006-0744

Thanks to Ernie Petrides and Asit B. Mallick for analysis and initial
patches.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

arch/x86_64/kernel/entry.S

@@ -180,6 +180,10 @@ rff_trace:
  *
  * XXX	if we had a free scratch register we could save the RSP into the stack frame
  *      and report it properly in ps. Unfortunately we haven't.
+ *
+ * When user can change the frames always force IRET. That is because
+ * it deals with uncanonical addresses better. SYSRET has trouble
+ * with them due to bugs in both AMD and Intel CPUs.
  */ 			 		
 
 ENTRY(system_call)
@@ -254,7 +258,10 @@ sysret_signal:
 	xorl %esi,%esi # oldset -> arg2
 	call ptregscall_common
 1:	movl $_TIF_NEED_RESCHED,%edi
-	jmp sysret_check
+	/* Use IRET because user could have changed frame. This
+	   works because ptregscall_common has called FIXUP_TOP_OF_STACK. */
+	cli
+	jmp int_with_check
 
 badsys:
 	movq $-ENOSYS,RAX-ARGOFFSET(%rsp)
@@ -280,7 +287,8 @@ tracesys:
 	call syscall_trace_leave
 	RESTORE_TOP_OF_STACK %rbx
 	RESTORE_REST
-	jmp ret_from_sys_call
+	/* Use IRET because user could have changed frame */
+	jmp int_ret_from_sys_call
 	CFI_ENDPROC
 
 /*